Pytorch搭建一个房价预测模型

以下是使用Pytorch搭建一个房价预测模型的步骤：

1. 导入必要的库

import torch
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

2. 加载数据集

data = pd.read_csv('housing.csv')

3. 数据预处理

# 将数据分为特征和目标
features = data.iloc[:, :-1]
target = data.iloc[:, -1]

# 数据标准化
features = (features - features.mean()) / features.std()

# 将特征和目标转换为NumPy数组
features = np.array(features)
target = np.array(target)
target = target.reshape(-1, 1)

4. 将数据集分为训练集和测试集

# 将数据集分为训练集和测试集，比例为80:20
from sklearn.model_selection import train_test_split

train_features, test_features, train_target, test_target = train_test_split(features, target, test_size=0.2, random_state=42)

5. 定义模型

class LinearRegression(torch.nn.Module):
    def __init__(self, input_size, output_size):
        super(LinearRegression, self).__init__()
        self.linear = torch.nn.Linear(input_size, output_size)

    def forward(self, x):
        return self.linear(x)

6. 实例化模型

model = LinearRegression(train_features.shape[1], 1)

7. 定义损失函数和优化器

criterion = torch.nn.MSELoss()
optimizer = torch.optim.SGD(model.parameters(), lr=0.01)

8. 训练模型

num_epochs = 1000

for epoch in range(num_epochs):
    # 将训练集转换为张量
    inputs = torch.from_numpy(train_features).float()
    targets = torch.from_numpy(train_target).float()

    # 清零梯度
    optimizer.zero_grad()

    # 前向传播
    outputs = model(inputs)

    # 计算损失
    loss = criterion(outputs, targets)

    # 反向传播和优化
    loss.backward()
    optimizer.step()

    # 每100个epoch输出一次损失
    if (epoch+1) % 100 == 0:
        print('Epoch [{}/{}], Loss: {:.4f}'.format(epoch+1, num_epochs, loss.item()))

9. 测试模型

# 将测试集转换为张量
inputs = torch.from_numpy(test_features).float()

# 预测房价
predicted = model(inputs).data.numpy()

# 计算均方根误差
from sklearn.metrics import mean_squared_error

mse = mean_squared_error(test_target, predicted)
rmse = np.sqrt(mse)
print('RMSE: {:.4f}'.format(rmse))

10. 可视化结果

plt.plot(test_target, label='Actual')
plt.plot(predicted, label='Predicted')
plt.legend()
plt.show()

完成以上步骤后，就可以使用Pytorch搭建一个房价预测模型了。